#AnthropicBansOpenClawFromClaude
Latest Update: #AnthropicBansOpenClawFromClaude
As of April 6, 2026, the developer community is adapting to the "soft ban" Anthropic placed on the open-source agent OpenClaw.
The Restriction: Effective April 4, 2026, users can no longer power third-party "harnesses" like OpenClaw using their flat-rate Claude Pro or Max subscription limits.
Infrastructure Costs: Anthropic justified the move by citing "outsized strain" on its systems, noting that agentic workflows in OpenClaw can consume over 30,000 tokens in a single session—far exceeding typical conversational use.
New Billing Model: While OpenClaw still works, users must now pay via direct API keys or a new "Extra Usage" pay-as-you-go tier, which can significantly increase monthly costs.
Strategic Rivalry: The move is seen as competitive, as it followed OpenClaw’s creator joining OpenAI and Anthropic launching its own rival agentic tools, Claude Code and Claude Cowork.
Refunds and Credits: To ease the transition, Anthropic is offering a one-month service credit ($20) to affected users, which must be claimed by April 17, 2026, or a full subscription refund for those who choose to cancel.